Path: utzoo!attcan!uunet!convex!convex.COM
From: patrick@convex.COM (Patrick F. McGehearty)
Newsgroups: comp.lang.fortran
Subject: Re: Random Number Functions--Generate versus Input
Message-ID: <108430@convex.convex.com>
Date: 8 Nov 90 20:28:56 GMT
References: <1990Nov8.170334.24579@athena.mit.edu>
Sender: usenet@convex.com
Reply-To: patrick@convex.COM (Patrick F. McGehearty)
Organization: Convex Computer Corporation, Richardson, Tx.
Lines: 13

In article <1990Nov8.170334.24579@athena.mit.edu> oliver@athena.mit.edu (James D. Oliver III) writes:
>I have a number (order of magnitude 10**6) of random number function values
>to generate ...
>  My question is, is it faster in general to regenerate the random function
> values every time (this part of the code can't be vectorized), or to just
> read the values in from a file?
It depends on how you read the numbers.

Using unformatted, direct I/O with large buffers into an array can be quite
efficient.  Reading each number with a separate READ from an ascii file
(thus requiring a string to real conversion) will be quite slow.
Whether the unformatted read is as fast as a random number function is
implementation dependent and should be easy to benchmark.